Project Summary Delirium in the intensive care unit (ICU) affects as many as 80% of critically ill, mechanically ventilated adults. Delirium is associated with substantial negative outcomes, including increased complications during hospitalization and long term effects on cognitive function and health status in ICU survivors. Sleep disturbances, iatrogenic/environmental factors, and biobehavioral factors may affect risk of delirium. This randomized clinical trial will test the effectiveness of a nonpharmacologic intervention to prevent delirium. The Family Automated Voice Reorientation (FAVoR) intervention uses scripted audio messages, recorded by the patient's family and played at hourly intervals during daytime hours, to provide information about the ICU environment to the patient; this ongoing orientation to the ICU environment through recorded messages in a voice familiar to the patient may enable the patient to more accurately interpret the environment and thus reduce risk of delirium. Increasing awareness of daytime by cuing patients during waking hours may also improve day/night orientation and nighttime sleep, further reducing risk of delirium. The primary specific aim of the project is to test the effect of the FAVoR intervention on delirium in critically ill, mechanically ventilated adults during hospitalization in the ICU. Secondary aims are to: (1) explore if the effect of FAVoR on delirium is mediated by sleep, (2) explore if selected biobehavioral factors may potentially moderate the effects of FAVoR on delirium, and (3) examine the effects of FAVoR on short term (immediately after ICU discharge) and long term (1 and 6 months after hospital discharge) outcomes, including cognitive function and patient-reported health status. A prospective, randomized, experimental design will be used to accomplish the specific aims. Subjects (n=178) will be randomly assigned to one of two groups (intervention or control) within 36 hours of intubation and ICU admission. The intervention group will receive FAVoR over a 5-day period, while the control group will not. Delirium free days, the primary outcome measure, will be assessed by Confusion Assessment Method (CAM-ICU) over the 5 day period. During the 5 day period, continuous polysomnography and actigraphy data and data related to iatrogenic/environmental and biobehavioral factors will be collected. After ICU discharge and at 1 and 6 months post-hospital discharge, follow up data (including cognitive function, patient reported health status, polysomnography, and actigraphy) will be collected. The effect of FAVoR on delirium free days will be tested using a Chi-square test; if there are unbalanced covariates, then covariance adjustment will be performed by using logistic regression that includes both the treatment assignment and the unbalanced covariates. Partial least square structural equation models will be used to analyze the relative importance and the pathways of secondary aim variables. FAVoR holds promise for reducing delirium in critically ill adults, improving sleep in the ICU, and secondarily benefitting cognitive function and health status in ICU survivors.